Open Access
Table 5
Comparative analysis between ANN and ANFIS.
| Sr. No. | Feature | ANN | ANFIS |
|---|---|---|---|
| 1. | Accuracy | 92–95% | 97–99% |
| 2. | Fault classification | Requires extensive training data | Efficient even with moderate training data |
| 3. | Fault detection speed | Moderate due to complex training weights | Faster due to hybrid learning approach |
| 4. | Adaptability | High for nonlinear systems but lacks interpretability | High with rule-based flexibility |
| 5. | Model interpretability | Low (black-box approach) | High (explainable with fuzzy rules) |
| 6. | Training time | Longer for large datasets | Relatively faster |
| 7. | Robustness to noise | Sensitive to noise in training data | Less sensitive due to fuzzy logic handling |
| 8. | Fault location precision | Accurate but slightly less consistent | Higher precision and consistency |
| 9. | Ease of implementation | Easier to implement in neural network frameworks | Requires defining fuzzy rules and membership functions |
| 10. | Generalization | Good with diverse training data | Very good due to hybrid learning |
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.